Agamotto — adversarial AI code/plan review with foresight. Opens the Eye of Agamotto banner on summon and does a quick foresight self-review, then sends the work to an external reviewer (Codex, or a fresh Claude reviewer subagent if codex is not installed or configured) while Claude fixes, looping until approved. Auto-detects plan/code/code-vs-plan mode. Invoke with /agamotto.
A rigorous "iterate until verified" working loop for any non-trivial build or fix. Use when the user prefixes a request with /dormammu, or asks you to be exhaustive, to "not stop until it's truly done", to self-critique hard, or to keep going until the solution is flawless. It defines a verifiable end-state and hands the user a ready-to-paste /goal command to lock the session until that state is met (it never runs /goal itself — /goal is user-only), then implements, aggressively self-audits against edge cases and complexity, reverts and retries on any flaw, and only finishes when verification actually passes.
Compile everything Claude did today across ALL sessions and repos into one dated teaching-guide markdown saved in "~/Daily Log", render it as a phone-readable PDF, and upload to Google Drive. Safe to run multiple times a day — it appends only newly-uncovered work to that day's single file rather than overwriting. Use when the user runs /wong, or says "log today's work", "end of day log", "daily log", "wrap up the day", or any similar end-of-day trigger. Accepts an effort level (light/medium/heavy, default heavy) to control token usage — e.g. "/wong medium" — and optionally a past date (YYYY-MM-DD).